cdcsis-package: Conditional Distance Correlation and Its Related Feature Screening Method

Description

Gives the conditional distance correlation and performs its based sure independence screening method, i.e., CDCSIS of Wen et al.(2014).

Details

Package: CDCSIS
Type: Package
Version: 1.0
Date: 2014-09-01
License: GPL(>=2)

The package cdcsis is used for calculate the conditional distance correlation and performs the related sure independent screening method. Details of the method can be found in Wen et.al. (2014).

Author(s)

Canhong Wen, Wenliang Pan, Mian Huang, and Xueqin Wang

Maintainer: Canhong Wen <wencanhong@gmail.com>

References

Canhong Wen, Wenliang Pan, Mian Huang and Xueqin Wang(2014). Conditional distance correlation sure independence screening for ultrahigh dimensional data. Submitted to Biostatistics.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
 set.seed(0)
 n <- 100 # sample size
 p <- 10 # dimensionality
 rho <- 0.5 # the correlation between pairwise predictors.
 Sigma <- matrix(rho, p, p)
 diag(Sigma) <- 1

 require(MASS)
 x <- mvrnorm(n,rep(0,p), Sigma)

 y <- x[,2] + x[,3] + rnorm(n)
 z <- x[,1]
 cdcsis(x,y,z,2)

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.